DLAI - Learning Platform Beta

Hi everyone, didn’t know where to post this so posted here.
In this lecture, we are using umap_plot which seems to be imported from utils library. I am unable to do the same while trying it out myself. Can anyone help me with the current syntax to do that ?
Python version : 3.10.12
umap-learn version : 0.5.3

Hi @Sohel_786,

This is from the short course Large Language Models with Semantic Search, so all related short courses queries should be posted in the Short Courses category.

And within the short courses category, this will go under Large Language Models with Semantic Search category.

For now, I have moved your query to the appropriate place, but for future do keep in mind where you post in order to get help promptly.

Cheers,
Mubsi

As for your query, I’m assuming you are trying to run the notebook on your own computer. But before doing that, did you download all of the relevant files accompanied by this notebook from the workspace ? The utils.py file especially ?

Hi there,
I had a similar issue as I was trying to copy the same code lines in my a new notebook on my laptop. I downloaded all the libraries that are needed for this exercise. But the same utils.py file does not have the umap_plot object. I checked umap folder in the site-package folder but could not find umap_plot. Also, the utils folder in the site-package folder does not have umap_plot as well.
So, I am not sure where to get this object from.
I am not sure how to download this notebook with all its dependencies from the course page. The notebook for lesson 2 about “Embedding” does not download!

Thanks and regards

Hi @haider.lawati,

Once the notebook loads do the following:

  1. Click File-->Open.... This will open your workspace in a new tab.
  2. Click the checkbox next to the notebook and then click Shutdown. This should change its colour from green to grey.
  3. Click the checkbox next to the notebook and then click Download.

If these instructions don’t help, let me know.

Hi Mubsi,
Yes it works now. Thank you so much.